Algorithms exist for compiling Fortran D for MIMD distributed-memory machines, but are significantly restricted in the presence of procedure calls. This paper presents interprocedural analysis, optimization, and code generation algorithms for Fortran D that limit compilation to only one pass over each procedure. This is accomplished by collecting summary information after edits, then compiling procedures in reverse topological order to propagate necessary information. Delaying instantiation of the computation partition, communication, and dynamic data decomposition is key to enabling interprocedural optimization. Recompilation analysis preserves the benefits of separate compilation. Empirical results show that interprocedural optimization is crucial in achieving acceptable performance for a common application.

This paper presents Fortran D, a version of Fortran enhanced with data decomposition specifications. It is designed to support two fundamental stages of writing a data-parallel program: problem mapping using sophisticated array alignments, and machine mapping through a rich set of data distribution functions. We believe that Fortran D provides a simple machine-independent programming model for most numerical computations. We intend to evaluate its usefulness for both programmers and advanced compilers on a variety of parallel architectures.

All rights reserved.

Fortran M is a small set of extensions to Fortran 77 that supports a modular approach to the design of message-passing programs. It has the following features. (1) Modularity. Programs are constructed by using explicitly-declared communication channels to plug together program modules called processes. A process can encapsulate common data, subprocesses, and internal communication. (2) Safety. Operations on channels are restricted so as to guarantee deterministic execution, even in dynamic computations that create and delete processes and channels. Channels are typed, so a compiler can check for correct usage. (3) Architecture Independence. The mapping of processes to processors can be specified with respect to a virtual computer with size and shape different from that of the target computer. Mapping is specified by annotations that influence performance but not correctness. (4) Efficiency. Fortran M can be compiled efficiently for uniprocessors, sharedmemory computers, distributed-m...

Massively parallel MIMD distributed-memory machines can provide enormous computation power. However, the difficulty of developing parallel programs for these machines has limited their accessibility. This paper presents compiler algorithms to automatically derive efficient message-passing programs based on data decompositions. Optimizations are presented to minimize load imbalance and communication costs for both loosely synchronous and pipelined loops. These techniques are employed in the compiler being developed at Rice University for Fortran D, a version of Fortran enhanced with data decomposition specifications. 1 Introduction It is widely recognized that parallel computing represents the only plausible way to continue to increase the computational power available to computational scientists and engineers. However, parallel computers are not likely to be widely successful until they are easy to program. A major component in the success of vector supercomputers is the ability of ...

Because of the complexity and variety of parallel architectures, an efficient machine-independent parallel programming model is needed to make parallel computing truly usable for scientific programmers. We believe that Fortran D, a version of Fortran enhanced with data decomposition specifications, can provide such a programming model. This paper presents the design of a prototype Fortran D compiler for the iPSC/860, a MIMD distributed-memory machine. Issues addressed include data decomposition analysis, guard introduction, communications generation and optimization, program transformations, and storage assignment. A test suite of scientific programs will be used to evaluate the effectiveness of both the compiler technology and programming model for the Fortran D compiler.

The success of large-scale parallel architectures is limited by the difficulty of developing machine-independent parallel programs. We have developed Fortran D, a version of Fortran extended with data decomposition specifications, to provide a portable data-parallel programming model. This paper presents the design of two key components of the Fortran D programming system: a prototype compiler and an environment to assist automatic data decomposition. The Fortran D compiler addresses program partitioning, communication generation and optimization, data decomposition analysis, run-time support for unstructured computations, and storage management. The Fortran D programming environment provides a static performance estimator and an automatic data partitioner. We believe that the Fortran D programming system will significantly ease the task of writing machine-independent data-parallel programs.

We present an overview of research at the CRPC designed to provide an efficient, portable programming model for scientific applications possessing both task and data parallelism. Fortran M programs exploit task parallelism by providing language extensions for user-defined process management and typed communication channels. A combination of compiler and run-time system support ensures modularity, safety, portability, and efficiency. Fortran D and High Performance Fortran programs exploit data parallelism by providing language extensions for user-defined data decomposition specifications, parallel loops, and parallel array operations. Compile-time analysis and optimization yield efficient, portable programs. We design an interface for using a task-parallel language to coordinate concurrent data-parallel computations. The interface permits concurrently executing data-parallel computations to interact through messages. A key notion underlying the proposed interface is the integration of F...

This paper describes a software architecture that allows image processing researchers to develop parallel applications in a transparent manner. The architecture's main component is an extensive library of low level image processing operations that can be run on distributed memory MIMD-style parallel hardware. Since the library has an application programming interface identical to that of an existing sequential image library, all parallelism is completely hidden from the user. In this paper we give an overview of all architecture components, and show how issues related to automatic parallelization and optimization are dealt with by the application of domain specic performance models. Results obtained for a realistic application indicate that model-based optimization of a wide range of imaging software indeed is possible. 1

Mentat is an object-oriented parallel processing system developed at the University of Virginia which has been ported to a variety of MIMD architectures. The computation model employed by Mentat is macro data-flow (MDF), a medium grain, scalable, data-driven computation model that supports both high degrees of parallelism and the object-oriented paradigm. A key aspect of the model is that it can be efficiently implemented. Inspired by data-flow, MDF retains the graph-based, data-driven, self-synchronizing aspects of data-flow. MDF address the shortcomings that data-flow exhibits when applied to distributed memory MIMD architectures by extending data-flow in three ways: (1) it is medium grain - actors are of sufficient computational complexity to amortize overhead costs, (2) program graphs are dynamically constructed at runtime - this permits dynamic function binding as required by the object-oriented paradigm and increases the average computation granularity, and (3) actors may maintai...